Framework for generating model-based system control parameters
First Claim
1. A control framework generating control parameters for controlling operation of a physical system, the control framework comprising:
- one or more embedded models each producing a model output corresponding to a different operating parameter of the physical system as a function of either of one or more operating values corresponding to operating conditions of the physical system and one or more of a number of solution parameters, objective logic producing a scalar performance metric as a function of the one or more model outputs, of one or more weight values and of one or more system performance target values, wherein the one or more model outputs defines a vector, Y, the one or more system performance target values defines a vector, YT, and the one or more weight values defines a vector, W, and wherein the objective logic is configured to determine a difference vector as a difference between the vectors Y and YT, and to determine the scalar performance metric as a vector inner product of the vector W and a function of the difference vector, objective optimization logic producing a number of unconstrained solution parameters in a manner that minimizes the scalar performance metric, and solution constraining logic determining the number of solution parameters from the number of unconstrained solution parameters in a manner that limits an operating range of at least one of the unconstrained control parameters, wherein the control parameters correspond to one of the number of unconstrained control parameters and the number of solution parameters, and wherein the solution constraining logic is further configured to produce at least one of the number of solution parameters as a function of at least one of the one or more system performance target values, and wherein the physical system is an internal combustion engine including an air handling system.
1 Assignment
0 Petitions
Accused Products
Abstract
A control framework generates control parameters for controlling operation of a physical system, and includes one or more embedded models each producing a model output corresponding to a different operating parameter of the system as a function of one or more system operating conditions and/or a number of solution parameters, objective logic producing a scalar performance metric as a function of the number model outputs and of one or more system performance target values, objective optimization logic determining a number of unconstrained solution parameters in a manner that minimizes the scalar performance metric, and solution constraining logic determining the number of solution parameters from the number of unconstrained solution parameters in a manner that limits an operating range of at least one of the unconstrained solution parameters. The control parameters may correspond to one of the number of unconstrained solution parameters or to the number of solution parameters.
-
Citations
51 Claims
-
1. A control framework generating control parameters for controlling operation of a physical system, the control framework comprising:
one or more embedded models each producing a model output corresponding to a different operating parameter of the physical system as a function of either of one or more operating values corresponding to operating conditions of the physical system and one or more of a number of solution parameters, objective logic producing a scalar performance metric as a function of the one or more model outputs, of one or more weight values and of one or more system performance target values, wherein the one or more model outputs defines a vector, Y, the one or more system performance target values defines a vector, YT, and the one or more weight values defines a vector, W, and wherein the objective logic is configured to determine a difference vector as a difference between the vectors Y and YT, and to determine the scalar performance metric as a vector inner product of the vector W and a function of the difference vector, objective optimization logic producing a number of unconstrained solution parameters in a manner that minimizes the scalar performance metric, and solution constraining logic determining the number of solution parameters from the number of unconstrained solution parameters in a manner that limits an operating range of at least one of the unconstrained control parameters, wherein the control parameters correspond to one of the number of unconstrained control parameters and the number of solution parameters, and wherein the solution constraining logic is further configured to produce at least one of the number of solution parameters as a function of at least one of the one or more system performance target values, and wherein the physical system is an internal combustion engine including an air handling system. - View Dependent Claims (3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13)
-
2. A control framework generating control parameters for controlling operation of a physical system, the control framework comprising:
-
one or more embedded models each producing a model output corresponding to a different operating parameter of the physical system as a function of either of one or more operating values corresponding to operating conditions of the physical system and one or more of a number of solution parameters, objective logic producing a scalar performance metric as a function of the one or more model outputs, of one or more weight values and of one or more system performance target values, wherein the one or more model outputs defines a vector, Y, the one or more system performance target values defines a vector, YT, and the one or more weight values defines a vector, W, and wherein the objective logic is configured to determine a difference vector as a difference between the vectors Y and YT, and to determine the scalar performance metric as a vector inner product of the vector W and a function of the difference vector, objective optimization logic producing a number of unconstrained solution parameters in a manner that minimizes the scalar performance metric, and solution constraining logic determining the number of solution parameters from the number of unconstrained solution parameters in a manner that limits an operating range of at least one of the unconstrained control parameters, wherein the control parameters correspond to one of the number of unconstrained control parameters and the number of solution parameters, and wherein the objective optimization logic includes solution selection logic responsive to a plurality of recent iterations of the scalar performance metric and to a corresponding plurality of recent iterations of the number of unconstrained solutions parameters to determine the control parameters as the one of the plurality of recent iterations of the number of unconstrained solution parameters having a corresponding one of the plurality of recent iterations of the scalar performance metric having a minimum magnitude with respect to remaining ones of the plurality of recent iterations of the scalar performance metric, and wherein the physical system is an internal combustion engine including an air handling system. - View Dependent Claims (14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39)
-
-
40. A method of generating control parameters for controlling operation of a physical system, the method comprising:
maintaining one or more embedded models each producing a model output corresponding to a different operating parameter of the physical system as a function of either of one or more operating values corresponding to operating conditions of the physical system and one or more of a number of solution parameters, producing a scalar performance metric as a function of the one or more model outputs, of one or more weight values and of one or more system performance target values, wherein the one or more model outputs defines a vector, Y, the one or more system performance target values defines a vector, YT, and the one or more weight values defines a vector, W, and wherein producing a scalar performance metric comprises determining a difference vector as a difference between the vectors Y and YT, and computing the scalar performance metric as a vector inner product of the vector W and a function of the difference vector, generating a number of unconstrained solution parameters in a manner that minimizes the scalar performance metric, determining from the number of unconstrained solution parameters the number of solution parameters in a manner that limits an operating range of at least one of the number of unconstrained solution parameters, and selecting as the control parameters one of the number of solution parameters and the number of unconstrained solution parameters, wherein the step of determining the number of solution parameters includes determining at least one of the number of solution parameters as a function of at least one of the one or more system performance target values, and wherein the physical system is an internal combustion engine including an air handling system. - View Dependent Claims (41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51)
Specification